--- title: "Deploy Discord Music Bot | (Just Updated) Lavalink Included, Health-Checked" description: "Lavalink audio server bundled and wired up, health-checked, plays at once" category: "Bots" url: https://railway.com/deploy/discord-music-bot-or-just-updated-lavali --- # Deploy Discord Music Bot | (Just Updated) Lavalink Included, Health-Checked Lavalink audio server bundled and wired up, health-checked, plays at once **[Deploy Discord Music Bot | (Just Updated) Lavalink Included, Health-Checked on Railway](https://railway.com/template/discord-music-bot-or-just-updated-lavali)** Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/discord-music-bot-or-just-updated-lavali/manifest.json - **Creator:** SuperSlowSloth - **Category:** Bots ## Template content ### lavalink - **Image:** ghcr.io/bon5co/discord-music-railway-lavalink:4.2.2 - **Health check:** /healthz - **Public domain:** Yes ### bot - **Image:** ghcr.io/bon5co/discord-music-railway-bot:4.6.7 ## Documentation # Deploy and Host a Discord Music Bot on Railway LavaMusic is an open-source Discord music bot: slash commands and a message prefix, queues, filters, playlists, autoplay, per-guild settings and 24/7 playback. It does not decode audio itself — like every serious Discord music bot it talks to **Lavalink**, a standalone audio server that resolves tracks and streams them into a voice channel. This template deploys both halves, already wired to each other. The only things you supply are your own Discord bot token and application ID. ## About Hosting a Discord Music Bot The thing that makes this class of bot awkward to host is that it is two programs, not one. Every other Discord music listing in this category deploys the bot alone and then asks the deployer for `LAVALINK_HOST` and `LAVALINK_PASSWORD` as blank, required deploy-form fields. There is no audio server anywhere in those templates, so the deploy succeeds, the bot logs in, the slash commands register — and every play command fails, because there is nothing to connect to. The deployer has to go and find a public Lavalink node, or host a second one somewhere else, before the thing they just deployed can play a single track. Nothing in the deploy surfaces that: no listing in this category publishes a healthcheck at all, so Railway reports SUCCESS either way. This template ships the audio server as a service in the same project: - **Lavalink 4.2.2** with the `youtube-source` plugin **baked into the image**, rather than downloaded from `maven.lavalink.dev` on every boot. A plugin fetch at start-up is a network dependency on someone else's Maven repository, and a coordinate that 404s is a hard boot failure. - The bot's `LAVALINK_URL` and `LAVALINK_AUTH` are filled in from the Lavalink service's own private domain and generated password. There is no audio configuration in the deploy form. - The Lavalink password is `${{secret(32)}}` — different on every deploy — and the server refuses to boot without one. **It is also health-checked, which stock Lavalink cannot be.** Every Lavalink HTTP route requires the `Authorization` header, so there is no endpoint a Railway healthcheck can call: an unauthenticated probe gets 401 and the deploy fails. This image puts a small nginx gateway in front of it that serves one unauthenticated route, `/healthz`, by making an *authenticated* call to Lavalink's own `/version`. A 200 there means the engine answered, not merely that a container is running. Everything else stays behind the password. Two other things this template gets right that are easy to get wrong on Railway: - **Railway's private network is IPv6-only.** A gateway listening on IPv4 only is unreachable from a sibling service over `RAILWAY_PRIVATE_DOMAIN`, so the gateway binds both. - **Railway mounts volumes owned by root**, and the bot image runs as an unprivileged user. The entrypoint repairs the mount, seeds the SQLite database from the image on first boot, and leaves it alone afterwards — verified across a real redeploy, where the second boot neither re-seeded nor re-repaired anything. The default search engine is **SoundCloud**, not YouTube. Unauthenticated YouTube playback from a datacentre IP is not dependable, and a bot whose default source fails intermittently is worse than one that is honest about it. Set `SEARCH_ENGINE` to `YouTube`, `Spotify`, `Deezer` or `Apple` if you have that working from your own deploy. ## Why Deploy a Discord Music Bot on Railway? Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it. By deploying this bot on Railway, you get both processes, the private network between them, the volume for the bot's database and a public HTTPS endpoint for the audio server without writing a compose file or renting a second box. ## Common Use Cases - **A music bot for your own Discord server**, with your own branding, prefix and language, and no third-party bot with access to your guild. - **A shared Lavalink node** — the audio server is also reachable over HTTPS with its generated password, so other bots you run can point at it. - **A base to build on** — LavaMusic is open source, so add commands or swap sources and redeploy. ## Dependencies for the Discord Music Bot - A Discord application with a bot token (`TOKEN`) and its application ID (`CLIENT_ID`), from the [Discord developer portal](https://discord.com/developers/applications). Enable the **Message Content** intent if you want the message prefix as well as slash commands. - A Railway volume for the bot's SQLite database. ### Deployment Dependencies - [LavaMusic](https://github.com/appujet/lavamusic) — the bot, pinned by digest. - [Lavalink](https://github.com/lavalink-devs/Lavalink) 4.2.2 and [youtube-source](https://github.com/lavalink-devs/youtube-source) 1.18.2. - Wrapper images and entrypoints: [bon5co/discord-music-railway](https://github.com/bon5co/discord-music-railway) ### Implementation Details After deploying, invite the bot with the OAuth2 URL for your application ID, granting the `bot` and `applications.commands` scopes. Join a voice channel and use `/play` (or `!play`, with the default prefix). `SEARCH_ENGINE`, `PREFIX` and `DEFAULT_LANGUAGE` are ordinary variables on the bot service and take effect on redeploy. The bot's database lives on the volume at `/data`. The Lavalink service is public so that it can be health-checked and reused; every route except `/healthz` requires the generated `LAVALINK_SERVER_PASSWORD`. Remove its domain if you would rather it stayed on the private network only. ## Similar templates - [Telegram JavaScript Bot](https://railway.com/deploy/5lRkWa) — A template for Telegram bot in JavaScript using grammY - [Cobalt Tools [Updated Sep ’26]](https://railway.com/deploy/cobalt) — Cobalt Tools [Sep ’26] (Media Downloader, Converter & Automation) Self Host - [Telegram Gateway](https://railway.com/deploy/railway-telegram-gateway) — Multi-bot Telegram webhook gateway with WebSocket event streaming Open this page in a browser: https://railway.com/deploy/discord-music-bot-or-just-updated-lavali